Everand Logo

Welcome to Everand!

  • Discover millions of ebooks, audiobooks, and so much more with a free trial

    Only $11.99/month after trial. Cancel anytime.

    Regular Algebra and Finite Machines
    Regular Algebra and Finite Machines
    Regular Algebra and Finite Machines
    Ebook207 pages3 hours

    Regular Algebra and Finite Machines

    By John Horton Conway

    4/5

    ()

    Read preview

    About this ebook

    World-famous mathematician John H. Conway based this classic text on a 1966 course he taught at Cambridge University. Geared toward graduate students of mathematics, it will also prove a valuable guide to researchers and professional mathematicians.
    His topics cover Moore's theory of experiments, Kleene's theory of regular events and expressions, Kleene algebras, the differential calculus of events, factors and the factor matrix, and the theory of operators. Additional subjects include event classes and operator classes, some regulator algebras, context-free languages, communicative regular algebra, axiomatic questions, the strength of classical axioms, and logical problems. Complete solutions to problems appear at the end.
    LanguageEnglish
    PublisherDover Publications
    Release dateSep 16, 2012
    ISBN9780486310589
    Regular Algebra and Finite Machines

    Related to Regular Algebra and Finite Machines

    Titles in the series (100)

    View More
    View More

    Related ebooks

    Mathematics For You

    View More
    View More

    Related podcast episodes

    Related articles

    • Article

      Are Neural Networks About to Reinvent Physics?

      Nov 21, 2019

      Can AI teach itself the laws of physics? Will classical computers soon be replaced by deep neural networks? Sure looks like it, if you’ve been following the news, which lately has been filled with headlines like, “A neural net solves the three-body p

      9 min read

    • Article

      Math’s Beautiful Monsters: How a destructive idea paved the way for modern math.

      Apr 3, 2014

      Much like its creator, Karl Weierstrass’ monster came from nowhere. After four years at university spent drinking and fencing, Weierstrass had left empty handed. He eventually took a teaching course and spent most of the 1850s as a schoolteacher in B

      10 min read

    • Article

      Physicists Attack Math’s $1,000,000 Question

      Apr 4, 2017

      Physicists are attempting to map the distribution of the prime numbers to the energy levels of a particular quantum system.

      4 min read

    • Article

      Math After COVID-19

      Apr 28, 2020

      Math After COVID-19
      4 min read

    • Article

      The Aesthetic Beauty of Math

      Jul 22, 2019

      The Aesthetic Beauty of Math
      5 min read

    • Article

      Five Veteran Scientists Tell Us What Most Surprised Them: Fifty years ago, who knew we’d learn to clone genes and find water on Mars?

      Sep 24, 2015

      Turn back the clock to 1965. Science appeared to be marching forward at an unrelenting pace. Biochemists had cracked the genetic code (how DNA translates into proteins), inspiring Life magazine to envision “superbabies with improved minds and bodies.

      7 min read

    • Article

      The Math of Living Things

      Jun 23, 2021

      It’s hard to argue with the famously authoritative Oxford English Dictionary, but its definition of physics as the “branch of science concerned with the nature and properties of non-living matter and energy” is incomplete because physics studies livi

      8 min read

    • Article

      The Computer Maverick Who Modeled the Evolution of Life

      Jul 11, 2019

      In 1953, at the dawn of modern computing, Nils Aall Barricelli played God. Clutching a deck of playing cards in one hand and a stack of punched cards in the other, Barricelli hovered over one of the world’s earliest and most influential computers, th

      10 min read

    • Article

      Actually, There Is a Time Like the Present

      Nov 30, 2017

      We’ve all heard it before: There’s no time like the present. Broadly speaking, of course, it means to “seize the opportunity right now,” or maybe in my case, to avoid procrastinating. From a psychological perspective, this makes a lot of sense. As hu

      4 min read

    • Article

      The Physicist’s New Book of Life

      Oct 21, 2020

      Who is Jeremy England? There are many answers to that question. He is a biochemistry graduate who became an MIT assistant professor in physics when he was 29 years old. He is an ordained rabbi. He is the grandson of Holocaust survivors. He is a desce

      11 min read

    • Article

      Life Is a Braid in Spacetime: How to See Yourself in a World Where Only Math Is Real.: How to see yourself in a world where only math is real.

      Jan 9, 2014

      “Excuse me, but what’s the time?” I’m guessing that you, like me, are guilty of having asked this question, as if it were obvious that there is such a thing as the time. Yet you’ve probably never approached a stranger and asked “Excuse me, but what’s

      7 min read

    • Article

      The Quantum Mechanics of Fate: How time travel might explain some of science’s biggest puzzles.

      Jan 30, 2014

      The objective world simply is, it does not happen,” wrote mathematician and physicist Hermann Weyl in 1949. From his point of view, the universe is laid out in time as surely as it is laid out in space. Time does not pass, and the past and future are

      6 min read

    • Article

      The Search For Wormholes

      Aug 11, 2022

      The Search For Wormholes
      11 min read

    • Article

      The Math That Takes Newton Into the Quantum World

      Feb 28, 2019

      In my 50s, too old to become a real expert, I have finally fallen in love with algebraic geometry. As the name suggests, this is the study of geometry using algebra. Around 1637, René Descartes laid the groundwork for this subject by taking a plane,

      9 min read

    • Article

      What Keeps Black Holes From Expanding Everywhere?

      Dec 10, 2018

      Black holes are somehow able to grow constantly without changing their size. Physics might finally be able to explain why.

      4 min read

    Related categories

    Reviews for Regular Algebra and Finite Machines

    4/5

    13 ratings0 reviews

    What did you think?

    Tap to rate

    Review must be at least 10 words

      Book preview

      Regular Algebra and Finite Machines - John Horton Conway

      Index

      __________________________________________

      Preface

      __________________________________________

      This book has been growing for some time. In 1960 I was particularly impressed by the papers of Kleene and Moore in Automata Studies (ed. Shannon and McCarthy), and decided to examine the axiomatics of Kleene’s algebra and to improve Moore’s bounds for lengths of experiments. Neither problem made much progress for some years, but then Michael Guy and I found the method of Chapter 2 for the Moore problem, and some time later I developed the treatment of the axiomatic problem described in Chapter 13, and spoke briefly about it at the 1966 International Congress of Mathematicians in Moscow. The matrix proof of Kleene’s theorem, and the differential calculus of events (Chapters 3 and 5), predate this. I have since learnt that the matrix formula was already known to P. J. Cleave in 1961, while J. Brzozowski has published several papers on the differential calculus in more recent years.

      I should like to thank in particular Andrew Glass, whose lucid notes put some sense into a lecture course I gave on these topics in Cambridge in 1966, and from which large parts of Chapters 0–6 and 12–13 have been copied almost verbatim, and my former research student Donald Pilling, whose work is the basis of Chapters 7–11. The idea of stressing the operator calculus is Pilling’s, and most of the results of Chapters 7–11 appear in some form or other in his PhD dissertation, even when this is not explicitly acknowledged in the text. In addition to his work on the operator calculus, Pilling seems to have given the first complete treatment of commutative regular algebra (Chapter 11). I should also like to thank Peter Rowat, whose earlier work helped to lay the foundations for some of the book, and Michael Guy for the contributions noted above.

      Not many references to the literature appear in the text, largely because I am not as familiar with the literature as I should be. For the same reason, most of the notation of the book is somewhat unorthodox, but I have taken some trouble to make it self-consistent, and hope that at least some of it will be found generally useful (excluding of course the bizarre names for particular operators). I hope that more than one reader will find the time to attack the various conjectures made, particularly in Chapters 6, 9, 11, 13 and 14. Not all of these can really be as hard as I have found them!

      __________________________________________

      CHAPTER ZERO

      Preliminaries to

      the Moore theory

      __________________________________________

      The main aim of this book is a presentation of Kleene’s algebraic theory of machines. To this end, we begin by formalizing the notion of machine. Then we present Moore’s identification theory of machines, which logically seems to precede the Kleene theory.

      Any actual machine is finite in the obvious sense that it has but a finite number of component parts, and so our theory will naturally be a theory of finite machines. Any actual machine is also to some extent unreliable, so that its behaviour is not always that which would be predicted from a knowledge of its design. However, we choose to ignore this unsavoury phenomenon, arguing that the deterministic theory is at once simpler and more interesting, and can in any case be made to include the probabilistic theory by the simple device of adding new input facilities accessible only to gremlins.

      Any machine worth the name will want to communicate with its environment by means of suitable input and output facilities. If the machine is deterministic, its output signals at any time will depend only on the state of the machine at that time, and the action of the machine at any instant will depend only on its state at that instant and the nature of any input signals it may receive.

      We there formalize the notion of finite machine (sequential machine, Moore machine, finite automaton) as follows:

      A finite machine M consists of:

      a finite set S = S(M) of objects called states

      a finite set I = I(M) of objects called inputs

      a finite set O = O(M) of objects called outputs

      a function t:S × I → S called the transition function

      a function o:S → O called the output function

      called the initial state.

      , … for states, a, b, c, … for inputs, and i, j, k, … for outputs, and we abbreviate t, aa.

      , continuously emitting output o), and it will remain in state a until it receives an input signal aa and emit output oa.

      If we think of the machine as a black box with push-buttons for inputs and lights for outputs, then the buttons must be spring-loaded so as to pop out after each pressing, and mechanically interlocked so that only one can be operated at any instant. Again, the electrical circuitry operating the outputs must be so designed that exactly one lamp is lit at any time. Other types of machine are readily accommodated by various coding processes – thus the formalized version of a machine with two input buttons a and b which can be operated separately or simultaneously, and two lamps i and j which can be lit or unlit independently of each other, will not have two inputs and two outputs, but rather three inputs (a, b, both) and four outputs (i, j, both, neither).

      Another popular formalization should perhaps be mentioned. A Mealey machine is defined just as a Moore machine, except that the output function is from S × I to O rather than simply from S to O. We think of such a machine as emitting its outputs at the instant of transition from one state to another, the output depending both on the previous state and the disturbing input. The theories of the two types are obviously equivalent, since any Mealey machine can be imitated by a Moore machine which has more states and ‘remembers’ the previous input, and a Moore machine can obviously be regarded as a Mealey machine in which the output happens to be independent of the input. We use the Moore version because it fits more simply into the Kleene theory.

      a and o), or graphically by means of its state diagram. The latter has a node for each state, marked with the name of that state and its output, and a directed branch marked a a . The initial state is marked by an ingoing arrow (→). When there are just two outputs 0 and 1 we call M a binary machinewith o) = 1. These remarks should be made quite clear by the example presented in Fig. 0.1.

      Fig. 0.1. A three-state machine and its state diagram.

      The Moore theory considers experiments on a finite machine M, under the supposition that the experimenter is forbidden to examine the state of M by any means other than observing its output. The aim is to extract information about the structure or state of M by applying various carefully chosen inputs and studying the resulting output-sequences. The following terminology is useful.

      A word is a finite (possibly empty) formal sequence of inputs. The length of the word is the length of this sequence, and the empty word 1 is the word of length zero. I* denotes the set of all words in inputs of Iis a state and w is the word ab … kw a)b)…)k obtained from a by applying the inputs a, b, …, k is accessible from only if there is a word w w , and accessible .

      are properties of the function f :I* → O(M) defined by f (w) = owindistinguishable if and only if f = f – that is to say, if and only if ow) = ow) for every word ware states of distinct machines M and M′ – all we need require is that M and M′ have the same console; that is to say, the same sets I and O.

      The engineer who wants a machine for some specific purpose will normally prefer the simplest machine that does the job. He will not usually approve of a multiplicity of parts with the same effect, and nor will he countenance the insertion of components with no function. These considerations lead naturally to the following notions.

      The quotient by indistinguishability

      For any machine M, we define a machine M ÷ with the same console by:

        (i) the states of M ) of M,

      a a

      (iii) the output function is defined by o) = o),

      .

      Theorem 1. M ÷ is indeed a machine, and any state of M is indistinguishable from the corresponding state of M ÷.

      Proof. M for every word w.

      The truncation by inaccessibility

      For any machine M, we define a machine M − with the same console by:

        (i) the states of M − are the accessible states of M,

       (ii) the transition and output functions of M − are the restrictions of those of M to the states of M −,

      (iii) the initial state of M .

      Theorem 2. M − is indeed a machine, and every state of M − is indistinguishable from regarded as a state of M.

      Proof. M a w has the same meaning in M − as in M.

      The point is, of course, that any two states of M ÷ are distinguishable, and any state of M − accessible. Figure 0.2 illustrates the two constructions.

      Fig. 0.2.

      We say that M is minimal if every state of M is accessible and every two distinct states of M are distinguishable. We say that M and N have the same behaviour if and only if their initial states are indistinguishable. The notion of isomorphism between two machines with the same console is defined in the obvious way.

      Theorem 3. M ÷− and M −÷ are minimal machines with the same behaviour as M. Minimal machines have the same behaviour if and only if they are isomorphic.

      Proof. The first statement follows easily from our previous theorems. If now M and N , we define a function from the states of M to those of N by fww for each word w. Then f is everywhere defined, since each state of M is accessible; well-definedw is the unique state of N w; and onto, since every state of N is accessible. It is therefore an isomorphism, since ow) = ow). The proof is completed by the remark that isomorphic machines have the same behaviour.

      We conclude that each machine has the same behaviour as some minimal machine, which is uniquely defined to within isomorphism. A possible snag in this otherwise very satisfactory state of affairs is that there might be no effective test for indistinguishability of states, and so perhaps no effective construction for this minimal machine. Fortunately, such a test is one of the first results of Chapter 1.

      __________________________________________

      CHAPTER ONE

      Moore’s theory

      of experiments

      __________________________________________

      The maintenance man in charge of a real machine M often has occasion to experiment with M. Sometimes he will be presented with M in some unaccustomed (and probably unknown) state and required to return it with a note specifying its new state. Occasionally he might suspect that some malfunction has transformed M into a new (and probably useless) machine M′, and will need to devise an experiment to determine whether this has in fact occurred. In both cases his method will probably be to apply certain inputs to the machine (each input depending on the previous outputs) until the resulting sequence of outputs in some sense contains the information he requires. The outcome of the experiment will be a

      Enjoying the preview?
      Page 1 of 1